Reverse engineer missing servomotor dataplate. BLDC??? Resolver.

I have an early CNC lathe with unmarked servos that may be BLDC units. There are no bosses for brushes on the motor case, and on inspection the interior of the motors are absolutely clean.





These servos are fitted with Kearfott control transmitters.

Their drives are CSR (Contraves) dex-syn units. I have found no documentation.



I am as yet uncertain as to the voltage of these amplifiers, but I think it may be ~150VDC.

What is the easiest way forward that retains the custom frame sized servos, and also, how to deal with the Kearfott resolver?

Thanks
.

The Mesa 7i49 card reads resolvers, and supplies analogue control voltage for the servo drives.

Both my mill and my lathe use resolvers on the axes and the spindles (using the 7i49)

Thanks Andy

Your Holbrook is inspiring and awesome!

I'm thinking to replace the CSR drives with something newer, Like Mesa 8i20 units.

I am puzzled by the physical parameters of my servomotors. They look like custom units. The only markings I've found on them is the identification of the Kearfott control transmitter. I'm guessing that they are BLDC.

The control on the lathe is an ancienct pre g-code thing that was programmed with thumbwheel switches. I think the CSR drivew may be PWM and not analog controlled. The schematics I have are not particularly clear on the motor + drive + feedback system. It is as if the manufacturer was trying to keep those details secret.

So as to interfacing I'd like to use 8i20 + 7i49 if possible.

Can the 7i49 work with a 3-phase control transmitter as well as a 2-phase resolver?

Also what power conditioning does the 8i20 want. The electrical cabinet is fitted with a healthy 3-phase rectifier but no filter bank.

Thanks
.

Best guest of the Kearfott device so far:

Input Voltage: 12.0 V
Input Freq: 2500 Hz
Primary: Stator
Input Current: 10.0 mA Max.
Output Voltage: 6.00 V Nom.
Trans. Ratio: 0.50 ± 5%

Would love to find a source for Kearfott docs.

No eBay APP ID and/or Cert ID defined in Kunena configuration

_jC wrote: The control on the lathe is an ancienct pre g-code thing that was programmed with thumbwheel switches. I think the CSR drivew may be PWM and not analog controlled.

That isn't a problem for LinuxCNC. You could probably test the theory with a software stepgen and parallel port.

Can the 7i49 work with a 3-phase control transmitter as well as a 2-phase resolver?

That will depend on the wiring. Can you test all the wires to each other with a multimeter and figure out the pairs?
I am using a funny resolver on the spindle of my lathe. primary is S1-S2 and S3-S4 and secondary is R1-R2 and R2-R3
(So two inputs and a centre-tapped output) but I managed to find a working combination with a bit of experiment.
It's easier if you have a signal generator and an oscilloscope. I have one of these and it has easily paid for itself:

Also what power conditioning does the 8i20 want. The electrical cabinet is fitted with a healthy 3-phase rectifier but no filter bank.

It needs DC, so a set of filter caps will be needed. They need to be low ESR and suitably rated for voltage too. I got some cheap from eBay, because the new price for the right sort of thing is rather high. (£100 per capacitir)

After slipping the protective covers off the servos it turns out that they are rather small for a contemporary machine.







For the effort required to retrofit this machine I've decided to go with something bigger.

.